Vol. 3,No.2, Feb., 1964
THE REACTION OF INDOLE AND TZ BACTERIOPHAGE
Keller, E. B., and Zamecnik, P. C. (195b), J. Biol. Chem. 221, 45.
Lengyel, P., Speyer, J. F,, and Ochoa, S. (1961), Proc. Nut. Acad. Sci. U . S. 47,1936. Loftfield, R. B., and Eigner, E. A. (1959), J. A m . Chem. Soc. 81, 4753.
Lowry, 0. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. (1951), J. Biol. Chem. 193, 265. Lubin, M. (1963), Biochim. Biophys. Acta 72,345. Luborsky, S. W., and Cantoni, G. L. (1962), Biochim. Biophys. Acta 61, 481.
McLean, J. R., Cohn, G. L., Brandt, I. K., and Simpson, M. V. (1958), J. Bwl. Chem. 233, 657. Matthei, J. H., Jones, 0. W., Martin, R. G., and Nirenberg, M. W. (1962), Proc. Nut. Acad. Sci. U . S. 48, 666. Mokrmch, L. C., and Manner, P. (1962), Biochim. Biophys. Acta 60, 415. Nakamoto, T., Conway, T. W., Allende, J. E., Spyrides, G. J., and Lipmann, F., (1963), Cold Spring Harbor Symp. Quant. Biol. 28, in press.
215
Nirenberg, M. W., and Matthei, J. H. (1961), Proc. Nat. Acad. Sci. U.S . 47,1588. Pennington, R. J. (1960), Biochem. J. 77,205. Sachs, H. (1957), J. Biol. C k m . 228, 23. Sager, R., Weinstein, I. B., and Ashkenazi, Y. (1963), Science 140,304.
Sanger, F. (1945), Biochem. J. 39, 507. Schneider, W. C. (1945), J. Biol. Chem. 161, 293. Siekevitz, P. (1952), J. Biol. Chem. 195, 549. Siekevitz, P. (1962), Methods Enzymol. 5 , 61. Simpson, M. V. (1962), Ann. Rev. Biochem. 31, 333. Volkin, E., and Cohn, W. E. (1954), in Methods in Enzymology, 1st ed., Glick, D., ed., New York, Interscience, p. 298.
Weinatein, I. B., and Schechter, A. N. (1962), Proc. Nut. Acad. Sci. U . S. 48,1686.
Winnick, R. E., and Winnick, T. (1960), J. Biol. Chem. 235, 2657.
Wool, I. G. (3963), Biochim. Biophys. Acta 68, 411. Yemm, E. W., and Cocking, E. C. (1955), Analyst 80, 209.
The Reaction of Indole and T2 Bacteriophage* L.
c. K A N N E RAND ~ L. M. KOZLOFFJ:
From the Department of Biochemistry, University of Chicago, Chicago, Illinois Received August 19,1963 Many aromatic compounds bind to bacteriophage T2 and prevent adsorption of the virus to its host cell. Indole and iodobenzene are the most active of a large number of substances tested. All active compounds are either T - or n-electron donors, and their ability to inhibit adsorption is correlated with their ability to form molecular charge-transfer complexes. Indole and other inhibitors cannot react with all of the alternative forms which T2 particles can assume in solution. The phage particles normally participate in a rapid equilibrium between a state in which they can adsorb to E . coli B (“active state”), and a state in which they cannot adsorb (“inactive state”). Indole reacts only with the nonadsorbing form of phage, and fixes the particles in the inactive form. The change from active to inactive phage results from a change in the tail-fiber configuration of the T2 particle from an extended to a nonextended state. This alteration of the fibers also affects the hydrodynamic behavior of T2. Bacteriophage T2H in the active state has a velocity sedimentation coefficient 10-15% smaller than it has in the inactive state. Anderson (1945, 1948) has shown that a medium containing L-tryptophan is required for the attachment of T4 bacteriophage to its host, Escherichia coli. The ability of a number of tryptophan derivatives and analogs to substitute for tryptophan was studied (Anderson, 1946), and it was concluded that the NH3+,COO-, and indole moieties of the molecule were all involved in the reaction with T4. Delbriick (1948) subsequently found that the activation of T4 by Ltryptophan was prevented if indole were also present, presumably by competition of the indole with the corresponding part of the tryptophan molecule.
* Aided by grants from the National Foundation, the John A. Hartford Foundation, Inc., the Dr. Wallace C. and Clara A. Abbott Memorial Fund of the University of Chicago, and a U. S. Public Health Service grant (GM-KB179-C4B).
t Predoctoral trainee on a Genetics Training Program (2G-90) sponsored by the National Institute of General
Medical Sciences, United States Public Health Service. This material was taken from a dissertation submitted in partial fulfillment of the requirements for the Ph.D. degree, University of Chicago, 1963, by Lee C. Kanner. Current address: Committee on Biophysics, University of Chicago, Chicago, Illinois. 3 Recipient of Public Health Service Research Career Award (l-K6-GM-179) from the National Institute of General Medical Sciences. On leave until September, 1964, at Department of Biology, University of California, La Jolla, California.
Bacteriophage T2 does not require tryptophan for adsorption, but indole was observed to inhibit adsorption of a strain of T2 known as T2H (Hershey and Davidson, 1951). We have found that indole will also inhibit T2L, another strain of T2. The reaction of T2 with indole (like that of T4 with tryptophan) affects the initial attachment of the phage to E. coli B. The indole reaction must involve the tail fibers of the phage, since these have been shown in two types of experiments to contain the sites of attachment to the host: (1) Isolated fibers were found to adsorb to host cells and retain their host specificity (Williams and Fraser, 1956); and (2) adsorption of T2 to E. coli B was found to be interrupted by a serum which was largely antifiber in activity (Franklin, 1961). The phage-indole or phage-tryptophan reaction also has been shown by immunological experiments to involve another component of the virus, the contractile sheath-protein. After reaction of phage with sheath antisera, T4 no longer has a tryptophan requirement for adsorption (Jerne, 1956), and indole no longer inhibits the adsorption of T4 or T2 (Jerne, 1956, Brenner etal., 1962). These observations on the role of both sheath and tail fibers have led to the hypothesis by Brenner et al. (1962) that: (a) T2 particles in their normal state and T4 particles which have reacted with tryptophan
216
Biochemistry
L. C. KANNER AND L. M. KOZLOFF
,
o o
l
"6.5
70
7.5
8.0
8.5
PH FIG.1.-Per cent inhibition of T2H adsorption as a function of p H . Phosphate or Tris buffer at ionic strength 0.15 was used. Standard procedures given under Methods and Materials were employed.
can adsorb because their tail fibers are extended and available for attachment to the host cell; and (b) "2 particles which have reacted with indole and T4 particles with no tryptophan added (or with tryptophan and indole added) cannot adsorb because their tail fibers are nonextended, i.e., bound to a component of the virus itself. The experiments described in this paper are concerned with three main aspects of phage-cofactor interaction: (a) the effect of the T2-indole reaction on the biological properties of the phage, i.e., on its ability to adsorb to E. coli B, and its susceptibility to Cd(CN),inactivation; (b) the effect of indole on the velocity sedimentation coefficient of T2H; and (c) the chemical nature of the binding of indole and other inhibitors to T2. METHODS AND MATERIALS Purification of Compounds Used.--The aromatic substances used (mostly obtained from commercial sources) were purified by recrystallization and/or distillation so that their spectra agreed with those given in the literature. Pyridyl pyridinium chloride * HC1 was decolorized in hot, dilute HC1 solution using acid-washed Norit A, evaporated to dryness, dissolved in warm ethanol, and precipitated by addition of ethyl ether. Benzothiophene was a gift of W. E. Parham. Bacteriophage Preparation.-Genetically homogeneous T2H was isolated from a single plaque from a stock of A. D. Hershey, and was prepared and assayed by standard procedures (Adams, 1959a). The T2L used was initially received from S. E. Luria several years ago. Anti-T2 serum was the gift of 0. Maaye. As used in the standard inhibition experiment described below it inactivated T2 a t a rate corresponding to K = 1.7min-1. Standard Inhibition Experiment: Determination of the Effectivenessof Various Compounds as Inhibitors.A concentrated stock of pursed bacteriophage T2H was diluted with a preincubation mixture of diluting %uid (0.025 M sodium diethylbarbiturate, p H 7, 0.125 M NaCl 0.001 M MgS04),or inhibitor dissolved in diluting fluid, to a concentration of 2 X 108 T2H/ ml. After several minutes, 1 volume (usually 0.1 ml) from this preincubation mixture was mixed with 19 volumes of a solution which had the same inhibitor
+
+
concentration and also contained 2 x 107 E . coli B/ml. Adsorption of the phage to the bacteria began at the time of mixing. The bacteria had been previously grown to about 5 X 108/ml in nutrient broth a t 37") sedimented and washed in diluting fluid four times, concentrated and added to the incubation mixture 1 minute before the phage was added from the preincubation mixture. The temperature of the preincubation and incubation mixtures was 30". A maximum of 8 % of the phage was adsorbed in 5 minutes. At 1 and 2 minutes after adsorption began aliquots were removed from the incubation mixture and put into solutions of T2 antisera. The sera irreversibly inactivated nonadsorbed phage particles so that they could not subsequently infect cells, but did not damage cells which had already been infected. After 10 minutes further dilutions were made, the samples were plated on nutrient agar with E. coli B, and the number of infected cells formed during the incubation period was determined from the number of plaques. A compound was tested as an inhibitor in this way at several concentrations, and the effect of a t least two concentrations of indole was determined a t the same time. Although there was a small variation from experiment to experiment in the amount of indole required for 50 yo inhibition, the relative effectiveness of an inhibitor was reproducible; i.e., the ratio: [inhibitorlsos, Inhlblt!on/ [indo1elao, r n h i b i t i o n remained constant. Known amounts of the relatively water-insoluble inhibitors, such as benzene and its derivatives, were added to the diluting fluid several hours before the addition of virus or bacteria. The tightly stoppered containers were shaken a t room temperature until the organic compounds were completely dissolved. It was possible to prepare solutions containing, for example, 5 x 10-4 M toluene (monomethyl benzene) and 10 -4 M xylene (dimethyl benzene). Determination of Association Constants with Chloranil as the Acceptor.-Association constants for the formation of complexes by aromatic electron donors with n-electron acceptor compounds were determined by measurement of the light absorption of the complexes (Benesi and Hildebrand, 1949) in the following manner. Solutions of the donor compound were made in 0.002 M chloranil in butyl ether. The donor concentration was varied (at least three concentrations were used) but was always in great excess over 0.002 M, the concentration of the chloranil. All solutions were made up under a red safelight; under these conditions the colors of the complexes formed remained stable for several hours. The absorbance of each chloranildonor solution was read a t a wavelength greater than 480 mk, where the 0.002 M chloranil does not absorb was appreciably. The association constant K,.,, calculated from: Kamoc
=
[complex ] ___ [donor]([chloranil],n,t,al - [complex])
-
~~
Substituting [complex] = absorbance1E, where e is the molar absorbance of the complex, and rearranging this expression, one obtains:
A graph of l/[donor] as a function of ljabsorbance gave a straight line, and its vertical intercept gave the negative of K,.,,,. Determination of Association Constants with Pyridyl Pyridinium Chloride.HCl as the Acceptor.---An aqueous medium was used as solvent: 0.025 M sodium diethyl-
+
+
barbiturate 0.125 M NaCl l o p RM MgSO,; this was the diluting fluid of the standard absorption experiments. The procedure was that described above for the butyl ether systems, except that the acceptor was present in excess, and its concentration was varied, while the donor was kept at a low fixed concentration. The association constant was then determined by the equation given above, but with the role of donor and acceptor reversed. The complexes were yellow, and absorbances were therefore read in the 40&500 mp region. Blanks were subtracted for the substantial absorbance of the pyridinium compound at these wavelengths. The donor was present in 0.02 M concentration, giving a blank value for absorbance of 0.00-0.02 at the wavelengths indicated; no correction was made for it. Velocity Sedimentation-Cwficient Determination.Velocity sedimentation coefficients for T2H were obtained using a Spinco Model E analytical ultracentrifuge and employing standard procedures given by Schachman (1959j.
1500
I
I
I
I
4
6
0
I
1250 -
1000
-
RESULTS Although the effect of indole was first observed in a system containing the bacterial host as well as the virus particles, it has since been established that indole reacts with the T2 particles themselves. Brenner et al. (1962) showed that the T2H-indole reaction protects T2H from Cd (CN)s- inactivation. Further, the sedimentation coefficient of T2H increased in the presence of indole (see below) a t concentrations which would also have been effective in inhibiting phage adsorption. Therefore inhibition is believed to be entirely due to the phage-indole reaction, not to an interaction of indole with E . coli B. Addition of indole to an incubation mixture reduced the number of infectious centers formed by T2H and its host cell by reducing the extent of initial attachment of the phage to the cell. No subsequent reaction necessary for infection (Kozloff, 1960) was affected; the number of infected cells formed during the incubation of T2H with a large excess of bacteria, plus the number of phage remaining in the supernatant after removal of the bacteria by centrifugation, remained constant in the presence or absence of indole, and equaled the total number of T2H particles added.
Effect of Indole on Adsorption and Inactivation of T2H and T2L Conditions Affecting the Inhibition Reaction of T2H.Indole and other inhibitors became more effective in preventing the adsorption of T2H to E . coli B when the incubation conditions were altered by one of the following procedures: increasing the divalent cation concentration, decreasing the monovalent cation concentration, lowering the temperature, and lowering the pH. All of these environmental changes also caused the adsorption rate to decrease, even with no inhibitor added. This is illustrated by the effect on adsorption and inhibition of the addition of M g + + to the incubation medium. AtpH 7,27O, ionic strength 0.15, and in the absence of divalent ions, the adsorption rate of T2H to the host cell was 2.6 X lo-@cm3min-I. The addition of 10 --3 M Mg caused the rate to drop to 1.6 x 10-9 cm3min-1. The effect of an inhibitor on the reaction rate was very small under the conditions of rapid adsorption, but greatly increased with the addition of Mg++. Table I shows the inhibition obtained with two inhibitors, indole and benzene, with or without Mg + + added. ++
"0
2
Incubation
8
Time
IO
(Minutes)
FIG.2.-Adsorption of T2H to E. coli B in the presence and absence of indole. Incubation mixture: 0.125 M NaCl, 0.025 M sodium diethylbarbiturate, 0.001 M MgSO,, HCl to p H 7; E. coli B a t 5 X 108/ml, T2H at 108/ml.
TABLE I OF T2H EFFECTOF ADDEDM g + +ON INHIBITION ADSORPTION" Inhibition Inhibitor 5 X 10-6M indole 5 x 10-3~ benzene
[Mg++l 0 10-3 M 0 10-3 M
( 70)
0 41 18 75
a Incubation mixture: 0.125 M NaC1, 0.025 M sodium diethylbarbiturate, M g + + as indicated, HC1 to p H 7; E . coli B at 2 X 107/ml; T2H a t 107/ml.
Figure 1 shows that lowering the p H similarly increased the effectiveness of a given concentration of indole. Kinetics of T2H Adsorption in the Presence of Indole.We had expected the reason for the decrease in the number of E . coli B infected with T2H in an indole solution to be a decrease in the rate of adsorption of the phage. However, this was not the case. Instead it was found (see Figure 2) that some of the phage particles (about 48% in the case illustrated) did not adsorb, regardless of the interval in which they were exposed to the cells, while most of the remainder adsorbed at about the same rate as they would have adsorbed in the absence of indole. The particles which could not adsorb were not inhibited by a component formed in the reaction medium during the 12 minutes of incubation. Figure 3 shows that when a new aliquot of T2H was added to the incubation mixture 6 minutes after the h t aliquot, the adsorption curve was repeated. Further, when an aliquot was removed from the incubation mixture in which adsorption had stopped, into an identical bacteria-indole mixture, adsorption still did not occur. The inability of some of the phage particles to adsorb in an indole solution
218
L.
c. KANNER AND L.
Biochemistry
M. KOZLOFF
TABLE I1 HETEROGENEITY OF T2H WITH RESPECT TO SENSITIVITY To Low INDOLE CONCENTRATIONS‘
Experiment
I
I1 I11 IV V
Indole Concentration ( X 10-j M)
Phage Concentration in Incubation Mixture (Per ml)
Incubation (1) Initial (2) With nonadsorbed phage from (1) Initial (2) With nonadsorbed phage from (1) Initial (2) With nonadsorbed phage from (3) With nonadsorbed phage from (1) Initial (1) Initial
1 1
0.7 0.7 0.7
(1)
(1)
(1) (2)
1.1 x 2.6 x 1x 2.4 x 8 X 8.2 x 2.1 x 1.2 x 3 x
108 103 108 103
Phage Unadsor bed after 5 Minutes
(%)
lo8
58 81 56 80 31
105 103 106 103
66 38 36
78
E. coli B a t 5 x 108,’ml in exper. I and 11; a t 2 X 109/ml in exper. 111, IV, and V.
I
“0
5
Incubation
IO
Time
15
(Minutes)
FIG.3.-Adsorption of T2H to E . coli B with 0, M, and 2 X 10 --6 M indole. Incubation mixture: 0.125 M NaC1, 0.025 M sodium diethylbarbiturate, 0.001 M MgSOd, HC1 to pH 7; E. coli B a t 5 X 108/ml, T2H a t 3 X 107/ml. was thus a property of these phage particles themselves, and not of the adsorption medium. It was not entirely surprising that almost all of the T2H was, in the presence of indole, divided into two stable populations, one containing phage particles which can adsorb (active phage), and the other phage particles which cannot adsorb (inactive phage). The fact that there is very little interconversion of these forms in the presence of indole is confirmed by the observation that T2H particles sediment in two distinct peaks in indole solution. However, when the indole concentration was altered the fraction of the phage which was active or inactive was instantaneously altered. Since no simple equilibrium existed between these forms, the response could not be explained by a “mass action” effect on a single equilibrium. The following alternative explanation was therefore considered. The number of inactive phage particles increased when the indole concentration increased be-
cause the phage particles varied in their sensitivity to indole. To investigate whether T2H are heterogeneous, as the foregoing implies, a “successive-incubation” experiment was used; phage particles which did not adsorb in one incubation were separated from the host cells and diluted out of indole, then put back into another indole-bacteria mixture to see if they were more susceptible than the original population to inhibition by the same concentration of indole. The results are summarized in Table 11. Using M indole, the fraction of inactive phage (phage in the supernatant) increased in experiment I from 58% in the first incubation to 81% in the second, and in experiment I1 from 56% to 80%. Using a lower indole concentration, three successive incubations were performed (experiment 111), yielding 31% inhibition in the first, 78% in the second, and 66% in the third. Control experiments (experiments IV and V) showed that the increase in inhibition obtained in the two final incubations of experiment I11 was not an artifact due to the use of a lower phage concentration. Since the phage stock used was prepared from a single-plaque isolation, it is unlikely that the heterogeneous response to indole found in these experiments is of genetic origin. The nature of the heterogeneity is not clear. Inhibition of T2L by Indole.-The adsorption of the strain T2L has been reported to be unaffected by indole (Adams, 1959b), and we have c o n h e d this in adsorption experiments using typical conditions for the incubation of phage with host cells. However, when these conditions were altered in the directions which had been found to increase the inhibition of TZH, inhibition of T2L could also be observed. For example, T2L adsorption was inhibited by indole a t temperatures in the range of 4-6”. Table I11 shons that inhibition also became greater as the pH was lowered: 2 X lod3 M indole was effective a t p H 6.7, M indole at p H 5.85. It should be and 2 X noted, however, that all experiments reported in this paper, except those given here, were performed using T2H. Indole Protection of T2H from Cd ( C N ) 3 -Inactivation. -We have found that indole is less effective in protecting T2H against Cd (CN)3- inactivation (Brenner et al., 1962) when Mg(N03)2is omitted from the standard Cd (CN),- inactivating solution (Kozloff and Lute 1957). The addition to the reagent of either 10-3 M Mg(N03)2or 10-3 M CaCL restored the protective action of the indole, an effect parallel to the divalent ion requirement for indole inhibition of adsorption. In Figure 4 the action of Mg++ on inactivation and protection is shown; Mg++did not significantly affect
Vol. 3, No. 2, Feb., 1964 EFFECTOF p H
ON THE
THE REACTION OF INDOLE AND Tg BACTERIOPHAGE
TABLE I11 INHIBITION OF T2L ADSORPTIONBY INDOLE^
PH
Indole 0
2 x 10-5~ 2 x 10-3~ 0 2 X 10-jM 2 x 10-3~ 0 2 X 10-jM 2 x 10-3~
5.58
6.7
7.55
Infectious Centers Formed in 1 Minute 30 10 1 632 514 117 652 724 513
-No
----
added Mg” with U 3 M Md’ added
Incubation mixture: phosphate buffer at ionic strength 0.1; [ M g + + ] = 0.67 X 10d4M; T = 21’.
the inactivation rate, but increased the protection afforded by indole to the phage.
Effect of Indole on the Sedimentation of T2H Velocity sedimentation coefficients were obtained for T2H a t p H 7.5 and 5.9, in phosphate buffer alone in buffer containing 3.3 x 10-5 M indole. Single peaks were obtained in both cases. At both pH values this concentration of indole cawed 100% inhibition of T2H adsorption to E . coli B. The sedimentation coefficient a t p H 7.5 increased in the presence of indole from 903 to 1012, while at p H 5.9, both with “ 1040) and without indole, the “fast” form ( s ~ ~ , = was obtained (Table IV). TABLE IV SEDIMENTATION OF T2H IN THE PRESENCE AND ABSENCE OF INDOLEG Indole
( x 10-5 M)
PH 7.5
0
5.9
3.3 0 3.3
Sedimentation Coefficient
IO
903 1012 1049 1038
+
Figure 5 shows the sedimentation of T2H a t pH 7.4 in buffered NaCl solution alone, or in 10-5 M or M solutions of indole. The comparable adsorption experiment for M and M indole gave 46% and 1 0 0 ~ oinhibition, respectively. A t a low indole concentration the T2H can be seen in the figure to have sedimented in two forms; in the absence of indole only the slower-sedimenting form was present = 945), and in the presence of sufficient indole to inactivate all of the phage for adsorption only the fastsedimenting form (s20,w= 1091)was present.
The Chemical Nature of the Binding of Indole to T2 Examples of inhibitors and noninhibitors of T2H adsorption are given in Table V. No common functional group could be identified as necessary for the reaction with the phage particle. Further, benzene, with no functional groups, was an effective inhibitor. It was noted that these aromatic inhibitors had in
20
Minutes FIG.4.-Effect of addition of M g + + and indole on Cd(CN)3--inactivation of T2H. The Cd(CN)3- reagent contained: 0.138 M Cd(N03)z,40.0 ml; 0.20 M NaCN, 86.4 ml; 0.30 M NaN03, 200 ml; 0.025 M Mg(N03)2,20 ml; when the Mg(NO8)*was omitted an additional 5 ml of 0.30 M NaNOs was added. The solutions were adjusted to p H 7.7 with H N 0 3 and diluted with H 2 0 t o a total volume of 500 ml. The phage were diluted 1-10 with the Cd(CN),- reagent.
TABLE V ACTIVITY OF AROMATICCOMPOUNDS AS INHIBITORS OF T2H ADSORPTION Inhibitors Single-Ring Compounds
Indole Derivatives
Benzene Toluene
D-
Bromobenzene Aniline
N-Acetyl indole 2-Methyl indole
or GTryptophan 5-Hydroxy tryptamine (serotonin)
(Ss0,w)
T2H used was a preparation dialyzed for 2 days against l o u 5 M MgSOa, adjusted t o p H 7. Sedi0.15 M NaCl mentation conditions: T2H at 2 X 1012/ml in phosphate buffer a t indicated pH, ionic strength 0.10; [ M g + + ] = 0.67 X loe6 M; T = 24.5’. a
219
Larger Fused-Ring Heterocycles Chlorpromazine Lysergic acid diethylamide tartrate (LSD-25)
Noninhibitors Pyrrole Pyridine
Purine Folic arid DPN
+
common the ability to act as electron-donors in the formation of molecular charge-transfer complexes (Mulliken, 1952). That is, each has the property of reacting reversibly with an electron acceptor to form a complex of the two molecules which is stabilized by the transfer of a nonbonding electron from an orbital of the electron-donor to an overlapping vacant orbital of an electron-acceptor molecule. Compounds are classified as n-donors or n-donors, according to whether the electron transferred is a n-electron of the donor ring-system or a localized p-electron of one atom of the donor molecule. If the formation of a chargetransfer complex is responsible for the inhibition reaction, it must also be true that in order to bind to the acceptor molecule on the phage only one orientation of the aromatic ring system of the donor, relative to the phage surface, is permitted. In this orientation the donor may act as a n- or n-electron donor, but only if the n- or p-orbital containing the electron to be transferred is oriented so that it overlaps the vacant orbital of the acceptor molecule on the phage. Pyridine
220
Biochemistry
L. C. HANNER AND 1.. M. K O Z M F F
!
FIG.S.-Sedimentation of T2H in the presence and absence of indole. Upper frames: lindolel = 0, 942; middle frames: [indole]= 1 0 ~ M, ~ , s, , ~ . ,=~ 949, 1122; lower frames: [indole] = lo-< M, = 1091. Sedimentation conditions: T2H at 2 X 10i3/rnl(by phosphorus determination); 0.09M NaCl phosphate buffer at p H 7.4 and ionic strength 0.01; [ M g + + l = 0.6 X 1 0 - j ~ ;T = 24.5".
s90.u =
+
S t e r d l y Samilarlnhibitorr of T2H Adsorption
INDOLE
BENZOFURAN
07 H
BENZOTHIOPHENE
a 2
m e" 07 INDAZOLE
BENZIMIDAZOLE
H
BENZOTRIAZOLE
OJ H
NAPTHALENE
00
cx; PURINE
QUINOLINE
00
H
FIG.6.4terically similar heterocyclic compounds which were tested t o determine their effectiveness as inhibitors of T2H. and ability t o act as r-electron donors.
and benzene provide an example of the effect of this steric requirement. Benzene is an inhibitor, and can be one only by virtue of s-electron donation, since it contains no mobile n-electrons. Pyridine, on the other hand, is not an inhibitor, though it is a good n-donor. Given the same "fit" to the phage particle as benzene, the p-orbital of its nitrogen atom is not oriented so as to permit charge transfer. Although r-electron dona-
tion is sterically permissible, pyridine is too weak a s-electron donor to be an inhibitor. To test the hypothesis that charge-transfer complexes are responsible for the binding of inhibitor molecules to T2, three series of compounds were examined to ascertain whether a correlation exists between their effectiveness as inhibitors of T2 adsorption and their effectivenessas electron donors. Comparison of Inhibit01 and Donor Properties of sElectron Donors.-Compounds which are sterically similar to each other were selected for comparison since it was found from experiments with substituted indoles that all side chains reduced inhibitor effectiveness. Figure 6 shows the nine compounde which were compared as adsorption inhibitors and as donors. Their inhibitor effectiveness was determined as a function of concentration (Figure 7j. Using the concentration for 50y0 inhibition as the criterion, the activity of a substance relative to indole was determined (Table VI). Extent of association with a common acceptor was measured for the same group of compounds, using chloranil as the acceptor, with butyl ether as solvent, for all the donor compounds which were soluble in this system a t sufficiently high concentration (Table VI j. These included indole, benzothiophene, benzofuran, naphthalene, and quinoline. Indazole is too polar to be measured in this way, hut an association constant was obtained in an aqueous system, using pyridyl pyridinium chloride.HC1 as acceptor. The aqueous system was also used for indole, the reference compound, and for benzofuran. Benzotriazole, benzimidazole, and purine are also too polar to use in the butyl ether system. However, in the aqueous system it was impossible to determine their association constants since the absorption spectra of the complexes (if any) were in the ultraviolet where the pyridinium compound itself absorbed very strongly. Inhibition of TZH Adsorption by n-Electron Donors: Halogenated benzenes can be expected to be excellent n-donors since their ionization potentials are low, rela~
Vol. 3, No. 2, Feb., 1964
FIG.7.-Inhibition
221
THE REACTION OF INDOLE AND T2 BACTEBIOPHAGE
of adsorption of T2H to E. coli B as a function of inhibitor concentration for the nine sterically similar heterocycles given in Figure 6.
TABLE VI ring by the acceptor will reduce the extent of a-electron A methyl substituent will not alter the disINHIBITOR EFFEXTIVENESS AND ASSOCIATION CONSTANTB transfer. FOR STERICALLY SIMILAR COMPOUNDS tance of closest approach to a benzene ring; this remains the same for methylbenzene as for benzene, 3.5 A. Relative But ethylbenzene has a larger distance of closest apInhibitor Relative Color in proach, about 5 A (MerrSeld and Phillips, 1958). EffectiveAssociation Butyl Therefore, complex formation is impeded for ethylCompound new Constant Ether benzene relative to methylbenzene, and the effect is 1.0 1.0 Purple Indole magnified if there is more than one ethyl group on the Benzothiophene 0.4 0.33 =t 0.06 Yellow aromatic ring. (Merrifield and Phillips 119581 found Napthalene 0.2 0 . 3 1 =t 0.03 Orange that the association constant for hexamethyl benzene 0.1 0.10 1 0.05 Yellow Benzofuran was fifty-one times greater than that for hexaethyl0 . 0 6 + 0.03b benzene using a common acceptor, tetracyanoethylIndazole 0.08 0.26 f0.05b ene). Quinoline 0.04 0 . 3 1 10.05 Yellow A comparison was made of the inhibitory effect on T2H adsorption of (a) methyl- and ethylbenzene, and Benzotriazole 0.006 (b) m-dimethyl- and mdiethylbenzene (Table VIII). Benzimidazole 0.001 Purine
